This population-based study used data from the GBD 2021 database, compiled by the Institute for Health Metrics and Evaluation (IHME). The study analyzed melanoma (ICD-10: C43) burden across all 50 U.S. states and the District of Columbia from 2010 to 2021, assessing incidence, prevalence, mortality, and DALYs. The Age-Standardized Incidence Rate (ASIR), Age-Standardized Prevalence Rate (ASPR), Age-Standardized Mortality Rate (ASMR), and Age-Standardized DALYs Rate (ASDR) were reviewed; the classifications of the Socio-Demographic Index (SDI), a standardized composite measure based on income per capita, average educational attainment, and total fertility rate, were used to assess socioeconomic status. DALYs (Disability-Adjusted Life Years) quantify the overall burden of disease by combining years of life lost due to premature mortality and years lived with disability. A higher DALY rate reflects a greater loss of healthy life years within a population.

While the SEER (Surveillance, Epidemiology, and End Results) program is a leading source of cancer incidence and survival data in the U.S., it covers only a subset of the population—approximately 48% as of recent estimates—and does not provide full nationwide or all-state coverage[1]. Additionally, SEER does not report disability metrics such as DALYs or include sociodemographic stratifiers such as the SDI. In contrast, GBD offers comprehensive, modeled estimates for all states, enabling standardized comparisons across time, geography, and socioeconomic strata—making it well-suited for analyzing disparities in melanoma burden on a national scale.

The methodology of the GBD was consistent with previous publications and the details provided in the Appendix. This study complies with the Guidelines for Accurate and Transparent Health Estimates Reporting (GATHER) statement (supplementary file1).

We reported age-standardized rates for each burden metric—incidence (ASIR), mortality (ASMR), prevalence (ASPR), and DALYs (ASDR)—using the direct method with the GBD 2021 world standard population. All rates were expressed per 100,000 population. We calculated the percent change between 2010 and 2021 to evaluate temporal trends. Linear regression was used for trend analysis; percentage changes in rates were reported with a 95% Uncertainty Interval (UI). Uncertainty intervals, used by the GBD study’s Bayesian framework, were defined as the 2.5th and 97.5th percentiles of 500 posterior draws from the framework and account for multiple sources of uncertainty[13]. To assess socioeconomic disparities, we stratified states by SDI and used one-way ANOVA to evaluate differences in melanoma burden metrics. Furthermore, to investigate potential drivers of geographic variation, we conducted an exploratory correlation analysis. State-level age-standardized incidence rates were correlated with demographic (median age, racial composition), environmental (UV irradiance), and structural (rurality, dermatologist workforce count, and location quotient) variables using Pearson’s or Spearman’s coefficients. The full methodology, data sources, and results for this exploratory analysis are detailed in Supplementary File S2. All statistical summaries and visualizations were generated using IBM SPSS Statistics for Windows, version 26 (IBM Corp., Armonk, N.Y., USA) and R, version 4.4.3 (R Core Team, 2025).

In 2021, there were 90,445 new cases of melanoma in the US (95% UI: 84,957–93,999), resulting in 9,996 deaths (9,185–10,436) and 263.5 thousand DALYs (247.5–280.6). ASIR was 17.33, ASPR was 139.06, ASMR was 1.75, and ASDR was 51.51 per 100,000. Between 2010 and 2021, melanoma incidence and mortality declined overall, with ASIR decreasing by 27.7% (male: 28.58%; female: 26.73%) and ASMR dropping by 24.89% (male: 26.6%; female: 23.49%). In 2021, the point prevalence of melanoma in the US was 0.15% while the Global prevalence was 0.03%. Refer to Table 1 for additional information.

Our exploratory analysis identified several state-level factors significantly associated with age-standardized melanoma incidence in 2021. A higher proportion of the White population showed a moderate positive correlation with incidence (r = 0.502, p < 0.001), whereas a higher proportion of the Black population demonstrated a moderate-to-strong negative correlation (r = − 0.591, p < 0.001). States with a higher median age also tended to have slightly higher incidence rates, though the correlation was weak (r = 0.277, p = 0.049). Among structural factors, a higher absolute number of dermatologists in a state was moderately negatively correlated with incidence (r = − 0.374, p = 0.032). No statistically significant associations were found for the proportions of Hispanic, Asian, American Indian/Alaska Native, or other racial groups. Likewise, environmental and other structural factors, including UV irradiance, rural population percentage, and dermatologist location quotient, showed no significant correlation with melanoma incidence (Supplementary Figs. 2–13).

From 2010 to 2021, the ASIR of melanoma declined in both sexes but remained consistently higher in males. In 2010, the ASIR among males was 30.95 per 100,000 (95% UI: 29.56–31.88), which declined to 22.11 (95% UI: 20.90–23.00) in 2021, representing a 28.6% reduction. Among females, the ASIR decreased from 18.46 (95% UI: 17.62–18.97) to 13.53 (95% UI: 12.75–14.07), a 26.7% decline.

The ASPR of melanoma was higher in males throughout the study period. In 2010, males had an ASPR of 243.53 per 100,000 (95% UI: 233.46–250.10), which declined to 171.02 (95% UI: 162.34–177.64) by 2021, a 29.8% decrease. In comparison, the ASPR for females declined from 155.63 (95% UI: 149.39–159.44) to 112.99 (95% UI: 107.34–117.32), a 27.4% reduction.

The ASDR demonstrated a similar male predominance and downward trend. In males, ASDR declined from 99.59 per 100,000 (95% UI: 94.78–105.13) in 2010 to 68.88 (95% UI: 64.97–73.00) in 2021, a 30.8% reduction. Among females, ASDR decreased from 50.42 (95% UI: 47.32–53.86) to 36.70 (95% UI: 34.28–39.40), a 27.2% reduction.

ASMRs declined in both sexes over the decade but remained markedly higher in males throughout. In 2010, the ASMR among males was 3.42 per 100,000 (95% UI: 3.23–3.52), compared to 1.49 (95% UI: 1.38–1.55) in females. By 2021, ASMR decreased to 2.51 (95% UI: 2.35–2.62) in males and 1.14 (95% UI: 1.04–1.20) in females, reflecting reductions of 26.6% and 23.5%, respectively. Despite these improvements, melanoma-related mortality remained substantially higher in males.

In 2021, melanoma incidence increased steadily with age, peaking in older adults (Fig. 2a-2). The highest incidence rate was observed in the 85–89 age group at 149.69 per 100,000 (95% UI: 117.94–166.33), followed by the 80–84 group at 121.79 (95% UI: 102.47–132.53) and the 90–94 group at 119.36 (95% UI: 91.46–135.59). Rates declined progressively in younger age groups, with the 65–69 group at 63.93 (95% UI: 60.27–67.34), and near zero among individuals under 20.

Prevalence also increased with age, with older adults experiencing the highest burden (Fig. 2b-2). The 85–89 age group had the highest prevalence at 845.71 per 100,000 (95% UI: 667.26–938.71), followed by the 80–84 group at 808.82 (95% UI: 677.15–882.56) and the 75–79 group at 725.97 (95% UI: 652.95–773.63). Prevalence declined in younger groups, including 523.78 (95% UI: 493.06–551.23) in the 65–69 group and 402.02 (95% UI: 380.02–425.22) in the 60–64 group. Prevalence was near zero among those under 20.

The melanoma burden, measured by DALY rate, rose sharply with age (Fig. 2c-2). The highest DALY rate was seen in the 85–89 age group at 318.99 per 100,000 (95% UI: 257.77–356.13), followed by the 90–94 group at 315.64 (95% UI: 242.23–358.43) and the 80–84 group at 294.19 (95% UI: 254.04–321.99). The 65–69 group had a DALY rate of 187.43 (95% UI: 176.53–199.30), while individuals under 20 had negligible rates.

Melanoma mortality increased consistently with age (Fig. 2d-2). The 95 + age group had the highest mortality rate at 33.30 per 100,000 (95% UI: 23.55–38.35), followed by the 90–94 group at 32.02 (95% UI: 24.62–36.17) and the 85–89 group at 26.24 (95% UI: 21.13–28.80). Mortality declined in younger age groups, reaching 0.66 (95% UI: 0.63–0.69) in the 35–39 group, and near zero among individuals under 20.

In 2021, the ASIR of melanoma varied widely across U.S. states (Fig. 3a-2). The highest rates were reported in Utah (21.66 per 100,000; 95% UI: 18.42–25.20), Colorado (21.60; 95% UI: 18.09–25.55), and New Hampshire (21.20; 95% UI: 17.73–24.94). The lowest ASIRs were observed in the District of Columbia (6.23; 95% UI: 5.24–7.31), Mississippi (12.72; 95% UI: 10.74–15.03), and Hawaii (12.83; 95% UI: 10.79–15.03).

Between 2010 and 2021, the national ASIR declined, though the magnitude of change varied by state (Supplemental Fig. 1a). Utah, which had the highest ASIR in 2010 (31.99; 95% UI: 29.44–34.40), experienced a 32.3% reduction. Similar declines were seen in Oregon (from 30.85 to 21.20; −31.3%) and Colorado (from 29.65 to 21.60; −27.2%). Several Midwestern states reported substantial reductions, including Minnesota (− 24.0%), Wisconsin (− 25.8%), Illinois (− 25.5%), and Michigan (− 23.3%). In the South, notable declines were observed in Texas (− 30.2%), Virginia (− 28.3%), and Alabama (− 29.1%). In the Northeast, Massachusetts (− 35.6%), Rhode Island (− 34.7%), and Connecticut (− 30.2%) also showed marked improvements. The District of Columbia, which had the lowest ASIR in both years, experienced a smaller reduction of 19.6% (from 7.75 to 6.23 per 100,000).

In 2021, the ASPR of melanoma also varied significantly by state (Fig. 3b-2). The highest ASPRs were observed in Colorado (175.65; 95% UI: 147.31–207.33), Utah (174.20; 95% UI: 147.82–203.23), Oregon (173.37; 95% UI: 146.23–208.39), New Hampshire (171.91; 95% UI: 144.34–201.22), and Arizona (171.05; 95% UI: 145.99–199.93). The lowest ASPRs were recorded in the District of Columbia (48.36; 95% UI: 40.77–56.42), Mississippi (97.63; 95% UI: 81.96–115.98), and Hawaii (107.15; 95% UI: 90.18–125.42).

All states experienced a decline in ASPR from 2010 to 2021 (Supplemental Fig. 1b). The largest reduction occurred in Massachusetts, from 252.77 (95% UI: 237.51–268.31) to 158.86 (95% UI: 131.66–188.85), a 37.2% decrease. Substantial reductions were also seen in Rhode Island (− 36.1%), North Dakota (− 33.7%), New York (− 33.3%), and Utah (− 33.6%). Other notable declines occurred in New Jersey (− 32.3%), California (− 29.2%), and Virginia (− 29.0%). In contrast, Mississippi (− 21.5%) and the District of Columbia (− 19.0%) had the smallest reductions in ASPR, reflecting persistent disparities.

In 2021, ASMR for melanoma varied across U.S. states (Fig. 3d-2). The highest ASMRs were observed in West Virginia (2.38 per 100,000; 95% UI: 2.05–2.74), Oklahoma (2.21; 95% UI: 1.89–2.56), Tennessee (2.20; 95% UI: 1.89–2.55), Kentucky (2.18; 95% UI: 1.86–2.55), and Utah (2.13; 95% UI: 1.80–2.47). States with the lowest ASMRs included New York (1.30; 95% UI: 1.09–1.51), Hawaii (1.10; 95% UI: 0.92–1.29), and the District of Columbia (0.87; 95% UI: 0.73–1.02).

Between 2010 and 2021, ASMR declined in all states, though reductions varied (Supplemental Fig. 1d). The largest percentage declines were seen in Massachusetts (− 30.8%), New York (− 30.1%), Rhode Island (− 29.9%), and New Jersey (− 29.5%). Utah, despite its high ASMR in 2021, showed a substantial decline of − 28.3%. States such as Texas (− 29.1%), Connecticut (− 28.3%), Oregon (− 27.6%), California (− 26.4%), and Virginia (− 25.3%) also exhibited notable reductions. In contrast, states with lower baseline ASMRs, like Hawaii (− 26.2%) and the District of Columbia (− 26.3%), experienced more modest improvements. These trends reflect broad progress in reducing melanoma-related mortality, though interstate disparities remain.

Males continue to exhibit a higher incidence and mortality than females. Although both sexes experienced improvements, the reductions were more pronounced among women, likely due to differences in photoprotective behavior, healthcare-seeking patterns, and occupational sun exposure [18–21]. A recent GBD analysis further attributed sex disparities to immune differences, hormone-related factors, and UV exposure patterns [22].. Liu et al. identified that men over 55 bore the highest mortality burden [7].

Age represents another critical demographic factor, with older adults bearing a disproportionate burden, reflecting cumulative UV exposure, immunosenescence, and reduced engagement in screening [23–25]. This was supported by our analysis (Supplementary file 2), where median state-level age showed a weak but statistically significant correlation with melanoma incidence (r = 0.28, p < 0.05). Social factors like marital status may influence outcomes; married individuals have lower melanoma mortality [26]. Educational gaps also persist. Leachman et al. found that few individuals in high-incidence states possessed sufficient melanoma literacy or confidence to perform self-skin examinations (SSE) [8].

Beyond sex and age differences, racial and ethnic background significantly influences melanoma outcomes across the U.S. While melanoma is far more common among non-Hispanic White populations (Supplementary file 2), individuals with skin of color, including Black, Hispanic, and Asian Americans, often face delayed diagnoses and worse prognoses [27]. States with a higher percentage of White population had a moderate positive correlation with melanoma incidence, whereas the percentage of Black population showed a moderate-to-strong inverse correlation. No statistically significant associations were observed for Hispanic, Asian, American Indian/Alaska Native, Native Hawaiian/Pacific Islander, or Multiple Races. Similar to our findings, Black men have a five-year survival rate of just 52%, compared to 75% among White men [28, 29]. These disparities stem not only from biological differences in melanin protection but also from limited awareness, a higher rate of different melanoma presentation (more aggressive and less immunotherapy responsiveness in acral lentiginous melanoma), and lower access to preventive care/screening in minority populations [30].

Geographic variation in ultraviolet (UV) radiation remains a central factor in melanoma epidemiology. A recent U.S.-focused analysis attributed approximately 91% of melanomas to UV exposure [31]. States located closer to the equator or at higher elevations, such as Colorado and Utah, are exposed to consistently higher UV indices throughout the year. These elevated UV levels, coupled with outdoor lifestyle patterns, may explain the persistently high melanoma incidence in these states [31].

However, the relationship between environmental UV exposure and melanoma incidence is more complex than expected. Geographic disparities remain substantial, with high-incidence states such as Utah, Colorado, and New Hampshire tending to have higher UV exposure and predominantly White populations, while states like the District of Columbia and Mississippi report lower rates, possibly due to higher proportions of individuals with darker skin or more urbanized environments [32]. Interestingly, this pattern contrasts with states like Hawaii, which report lower overall melanoma rates despite high ambient UV exposure. Notably, SEER data indicate that Native Hawaiian populations have a disproportionately high melanoma risk, suggesting that individual-level susceptibility, behavioral patterns, or access to care may play a significant role beyond environmental exposure alone [33]. It is also important to note that this may reflect that ambient UV is not equivalent to actual personal exposure, which is shaped by altitude (e.g., Utah’s high elevation), occupation, recreational patterns, and protective behaviors at the individual level. States with similar absolute UV levels can have markedly different burdens due to these modifying factors.

Surprisingly, our results showed that higher melanoma age-standardized estimates are not statistically correlated with UV index (Supplementary Table 2). It could be due to variability in skin cancer awareness throughout the population, including avoidance of UV exposure and the use of sunscreen; this needs to be addressed in future studies. This finding suggests that behavioral clustering and socioeconomic factors can override environmental protection, as illustrated by a recent geospatial analysis in New England. Counties like Grand Isle, Windsor, and Grafton had elevated melanoma rates despite lower ambient UV exposure, driven by older age, income, and proximity to tanning facilities [34].

The temporal dimension adds another layer of complexity to environmental determinants. The long latency between UV exposure and melanoma onset also complicates temporal analyses. Elevated rates in Massachusetts and Rhode Island may reflect childhood UV exposure from earlier decades, before widespread adoption of sunscreen and public health interventions [35]. The climate in these states is also associated with intermittent, high-intensity seasonal exposure.

Approximately 15% of Americans reside in rural counties, yet these areas remain chronically underserved by dermatologic services. Nationwide analyses have documented a persistent shortage of dermatologists in rural regions, where specialists tend to cluster in urban centers [36]. This imbalance creates significant access barriers: many rural patients must travel long distances for evaluation, contributing to diagnostic delays and, ultimately, higher melanoma mortality [37]. These access gaps may partially explain the elevated burden observed in low-SDI and Appalachian states.

The dermatologist shortage extends beyond rural areas. Even outside rural regions, overall dermatologist availability remains limited relative to national demand. Recent workforce estimates suggest there are only 3–4 dermatologists per 100,000 Americans [38]. While the number of providers has grown over the past decade, this growth has not kept pace with increasing patient volume, especially among aging populations and in high-risk states. A 2017 national report concluded that the current training pipeline may be insufficient to meet future demand[38], particularly in areas with already limited infrastructure.

It is notable that in our analysis, the dermatologist workforce count was one of the only three variables to show a statistically significant relationship with incidence.(supplementary 2) The absence of an association with dermatologist location quotient suggests that absolute specialist numbers, rather than their proportion relative to the population, may better capture aspects of access, visibility, and referral dynamics that influence case detection.

The relationship between healthcare infrastructure and outcomes, however, reveals additional complexity. Access to dermatologic care is another critical determinant. Hopkins et al. found that states with higher dermatologist density were associated with improved survival in univariable analysis, though this relationship could not be confirmed in multivariable models due to data limitations [39]. States like Utah and Massachusetts, with robust medical infrastructure, such as higher physician availability or cancer center service, might reflect both better detection and more complex case referrals. While others with more limited healthcare resources, like Kentucky and West Virginia, continue to report high mortality despite lower incidence [39].

In contrast to earlier reports of rising incidence of melanoma in US with stable mortality[40, 41].—often interpreted as overdiagnosis of indolent lesions—our data show parallel declines in incidence, prevalence, and mortality. This pattern supports a true reduction in melanoma burden rather than diagnostic inflation. The timing is consistent with the lag expected for large-scale prevention efforts, such as public UV protection campaigns initiated in the 1980s, to translate into measurable decreases in incidence and mortality. These findings highlight the long-term value of sustained prevention, combined with access to specialist care, in reducing both new cases and deaths from melanoma.

These structural disparities align with broader socioeconomic patterns. Finally, our findings align with broader global trends. High-SDI states in the U.S. outperform lower-SDI counterparts on mortality and DALY reduction, echoing similar international patterns [7, 42].

This study has several limitations that must be considered when interpreting our findings. First, the GBD framework relies on statistical modeling and imputation, which may introduce bias in regions with limited data availability. The absence of cancer staging information and individual-level ethnicity data in GBD datasets limits our ability to assess disease severity patterns or mortality differences in specific vulnerable populations.

Second, and critically, our correlation analysis was exploratory and conducted at the state level; findings are hypothesis-generating and cannot be used to infer individual-level risk. (Supplementary File 2)

Third, we used covariate data from different years (e.g., UV for 2020–2023), which may introduce minor temporal mismatch. Because the correlation analyses were exploratory and uncorrected for multiple comparisons, p < 0.05 findings should be interpreted cautiously and confirmed in future studies..

Fourth, ambient UV exposure measurements may not accurately reflect personal UV exposure, which is modified by individual behaviors, occupational patterns, and protective practices. The lack of correlation between state-level UV index and melanoma incidence likely reflects this disconnect between environmental and actual exposure.

Finally, the long latency period between UV exposure and melanoma development complicates temporal analyses. Current melanoma rates may reflect exposure patterns from decades prior, before widespread adoption of sun protection measures. Despite these limitations, our findings provide valuable insights into melanoma burden patterns and disparities across the United States, highlighting areas warranting targeted intervention and further investigation.